Transfer Learning for Named Entity Recognition of Classical Latin through LLM Prompting
Quick Answer
This paper shows that Team uOttawa achieved top results in Named Entity Recognition for Classical Latin using LLMs gemini-2.5-pro and claude-sonnet-4-5.
Quick Take
The system excelled in both coarse and fine-grained NER tasks, outperforming all submissions with the best scores across evaluation metrics. This demonstrates the potential of cross-lingual transfer learning for underrepresented ancient languages.
Key Points
- Team uOttawa ranked first in both coarse and fine-grained NER tasks.
- Used gemini-2.5-pro and claude-sonnet-4-5 for prompt engineering.
- Achieved best scores across all evaluation metrics and regimes.
- Demonstrated effective cross-lingual transfer learning for Classical Latin.
- Contributed to the EvaLatin 2026 research initiative.
Paper Resources
📖 Reader Mode
~2 min readAbstract:With the increase in digitized resources of Classical Latin texts and modern breakthroughs of Large Language Models (LLMs), I contribute to ancient language research by participating in EvaLatin 2026. This paper describes Team uOttawa's system description and results for the Named Entity Recognition (NER) shared task. The task is divided into two subtasks: coarse-grained NER with 11 classes and fine-grained NER with 28 classes, each evaluated under strict and fuzzy regimes. Through prompt engineering of commercial LLMs gemini-2.5-pro and claude-sonnet-4-5, I show that the underrepresented ancient Latin language can take advantage of cross-lingual transfer learning by using advancements made by the wider LLM development community. Overall, the methods discussed in this report demonstrate very strong results, placing first in both NER subtasks and achieving the best scores across all evaluation metrics and regimes among all submissions.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2608.04015 [cs.CL] |
| (or arXiv:2608.04015v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2608.04015 arXiv-issued DOI via DataCite |
|
| Journal reference: | EvaLatin (LT4HALA@LREC), ELRA, May 2026, Palma De Majorque, Spain |
Submission history
From: Callum Chan [view email]
[v1]
Tue, 26 May 2026 23:37:49 UTC (62 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CL
See more →TriAgent: Divergence-Aware Committees for Cost-Efficient Financial Sentiment Analysis
TriAgent introduces a cost-efficient multi-agent system for financial sentiment analysis, combining VADER, FinBERT, and Qwen2.5. It achieves an F1 score of ~0.87 with significant savings of $9.3M/year at a 10M-user scale compared to GPT-4o-mini, while also detecting hallucinations with an AUC of 0.90.